Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/Pierry/harness-kitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/pierry/harness-kit/pr-monitor)<a href="https://agentmods.dev/commands/pierry/harness-kit/pr-monitor"><img src="https://agentmods.dev/badge/commands/pierry/harness-kit/pr-monitor/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/pierry/harness-kit/pr-monitor"><img src="https://agentmods.dev/badge/commands/pierry/harness-kit/pr-monitor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00028 | $0.00601 |
| Opus 5 | $0.00014 | $0.00300 |
| Sonnet 5 | $0.00006 | $0.00120 |
| Haiku 4.5 | $0.00003 | $0.00060 |
Grade A, and why
pr-monitor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Monitor active PR until it merges. Polls with backoff: 3min → 6min → 12min → 24min → 30min cap, escalating after 5 attempts at each rung.
Runs in active session via ScheduleWakeup. Closing session ends monitor.
Flow
-
Read state from
.claude/.pr-monitor-state.json.- Missing → first run. Detect PR from current branch:
No PR or state already MERGED → exit silently (nothing to watch). Otherwise initialize state:gh pr view --json number,url,headRefName,state
Script prints next-delay in seconds..claude/scripts/pr-monitor.py init <number> <url> <branch> - Present → use
state.pr_number.
- Missing → first run. Detect PR from current branch:
-
Check merge status:
gh pr view <pr_number> --json state -q .state -
If
MERGED:- Print one line:
PR #<n> merged → <url>. Pipeline state cleared. Start next: /product-manager:run - Clear monitor state:
.claude/scripts/pr-monitor.py clear - Clear pipeline state:
.claude/scripts/pipeline.py clear - Do not call ScheduleWakeup. Done.
- Print one line:
-
If
CLOSED(not merged):- Print:
PR #<n> closed without merge. Monitor stopped. - Clear monitor state only. Done.
- Print:
-
If
OPEN:- Bump attempt counter, get next interval:
Script prints next-delay in seconds (e.g..claude/scripts/pr-monitor.py bump180,360,720,1440,1800). - Read state with
.claude/scripts/pr-monitor.py readforcurrent_interval_minandtotal_attemptsto include in user-facing line. - Print one line:
PR #<n> still open (attempt <total>, next check in <m>min). - Schedule next wake:
- Load
ScheduleWakeupvia ToolSearch:select:ScheduleWakeup - Call with
delaySeconds=<next>,prompt="/sse:pr-monitor",reason="watching PR #<n> for merge, interval <m>min".
- Load
- Bump attempt counter, get next interval:
First invocation (right after /sse:pr)
No state file yet. Initialize then schedule first wake at 3min (180s). Print:
PR monitor armed for #<n>. First check in 3min, escalates to 30min cap.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 57 lines · 28 tokens per session scan A a1a8229dd7cd
pr-monitor is a command published in the GitHub repository Pierry/harness-kit (3 stars, last pushed 1mo ago), licensed MIT. It adds 28 tokens to every session and 601 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
branch
Show Claude Code costs per git branch / PR.
changelog
Generate a changelog entry from git commits since the last tag using conventional commit format.
deploy
Deploy application with pre/post-deploy checks.
pull-repos
Pull all repos (parent + marketplace clones + configured project repos).
auto-claude
Invoke Auto-Claude autonomous coding framework for complex feature implementation.
git-ops
Git + worktree orchestrator entry point. /git-ops --landall surveys every branch/worktree and batch-lands the ones that are done; bare /git-ops runs a status survey. Thin router over the git-ops skill.